CN109376257A - Tealeaves recognition methods based on image procossing - Google Patents

Tealeaves recognition methods based on image procossing Download PDF

Info

Publication number
CN109376257A
CN109376257A CN201811245694.XA CN201811245694A CN109376257A CN 109376257 A CN109376257 A CN 109376257A CN 201811245694 A CN201811245694 A CN 201811245694A CN 109376257 A CN109376257 A CN 109376257A
Authority
CN
China
Prior art keywords
tealeaves
tea
color image
rgb
yellow leaf
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201811245694.XA
Other languages
Chinese (zh)
Inventor
李志强
甘密
杨洪涛
汪飞
夏先春
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guizhou Electromechanical Research & Design Institute
Original Assignee
Guizhou Electromechanical Research & Design Institute
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Guizhou Electromechanical Research & Design Institute filed Critical Guizhou Electromechanical Research & Design Institute
Priority to CN201811245694.XA priority Critical patent/CN109376257A/en
Publication of CN109376257A publication Critical patent/CN109376257A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/90Determination of colour characteristics
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00Image enhancement or restoration
    • G06T5/001Image restoration
    • G06T5/002Denoising; Smoothing
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/0002Inspection of images, e.g. flaw detection
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/10Segmentation; Edge detection
    • G06T7/155Segmentation; Edge detection involving morphological operators

Abstract

The tealeaves recognition methods based on image procossing that the invention discloses a kind of, (A1) obtain the RGB original color image of sample tealeaves, tea stalk and yellow leaf;(A2) RGB original color image is made into filtering and noise reduction pretreatment;(A3) hsv color space will be transformed into from RGB color by pretreated RGB original color image form HSV color image;(A4) the tone H of tealeaves, tea stalk and yellow leaf HSV color image, the characteristic parameter of saturation degree S, lightness V are extracted;(A5) according to the characteristic parameter of the tone H of tealeaves, tea stalk and yellow leaf, saturation degree S, lightness V, the upper and lower limit value of respective tone H, saturation degree S, lightness V are determined, so that it is determined that the respectively value range of tone H, saturation degree S, lightness V;(A6) by the tea dry sorting control system of determining value range typing tea dry sorting equipment, database is constituted;B, the acquisition of characteristics of objects parameter;C, the matching of characteristic parameter.This method can rapidly sort out the tea stalk in tealeaves, yellow leaf.

Description

Tealeaves recognition methods based on image procossing
Technical field
The present invention relates to a kind of tealeaves recognition methods, especially a kind of tealeaves recognition methods based on image procossing.
Background technique
Image processing techniques is the development and an important applied field that is mature and developing rapidly with computer, It is simulation eye recognition mechanism, using color as a kind of technical method of main characteristic parameters.With computer image processing technology The intelligent operation of information collection and processing can be achieved, and lossless, quick, real-time monitoring is carried out to identification object, improve and calculate The precision and efficiency of method are widely used to production and the manufacture field of modern agriculture.
In recent years, many scholars are dedicated to the research of tealeaves identification, and some scholars utilize the color characteristic pair of color image Tealeaves tender leaf is identified;Another part scholar identifies the finished tea after processing using image processing techniques. In the identification to finished tea, someone identifies tealeaves using near-infrared spectrum technique, and someone is known using BP neural network technology Other tealeaves, there are also groups of people, and tealeaves is identified using support vector machines, and recognition effect is preferable, but these methods are that identification is different The method of type tealeaves carries out identification to same tealeaves to screen tealeaves, tea stalk and the yellow leaf in same tealeaves Method out rarely has people's research.
Summary of the invention
The tealeaves recognition methods based on image procossing that the object of the present invention is to provide a kind of.This method can be to tealeaves It is monitored, lossless, rapidly can be sorted out the tea stalk in tealeaves, yellow leaf in real time.
Technical solution of the present invention: a kind of tealeaves recognition methods based on image procossing, this method include building for database The vertical, acquisition of characteristics of objects parameter and the matching of characteristic parameter, specific steps include:
A, the foundation of database
(A1) the RGB original color image of sample tealeaves, tea stalk and yellow leaf is obtained;
(A2) RGB original color image is made into filtering and noise reduction pretreatment;
(A3) HSV color space will be transformed into from RGB color by pretreated RGB original color image form HSV Color image;
(A4) the tone H of tealeaves, tea stalk and yellow leaf HSV color image, the characteristic parameter of saturation degree S, lightness V are extracted;
(A5) according to the characteristic parameter of the tone H of tealeaves, tea stalk and yellow leaf, saturation degree S, lightness V, respective tone is determined H, the upper and lower limit value of saturation degree S, lightness V, so that it is determined that the respectively value range of tone H, saturation degree S, lightness V;
(A6) by the tea dry sorting control system of determining value range typing tea dry sorting equipment, database is constituted;
B, the acquisition of characteristics of objects parameter
Tealeaves to sub-sieve is put into tea dry sorting equipment, in tealeaves transmission process, step (A1) is repeated to (A4), passes through The characteristic parameter of tea dry sorting control system acquisition object;
C, the matching of characteristic parameter
The characteristic parameter for the object that will acquire is matched with the value range of tealeaves, tea stalk and yellow leaf in database, is confirmed Object belongs to tealeaves, tea stalk, yellow leaf or other.
The RGB original color image acquisition methods of tealeaves recognition methods above-mentioned based on image procossing, object are, will be at Leaf of sampling tea is put on a vibrator platform, and as vibrator vibrates, finished tea is 30 ° -70 ° of skewed slot along gradient Even drop down obtains RGB original color image by CCD color camera into the shooting area of CCD color camera.
Tealeaves recognition methods above-mentioned based on image procossing, in the RGB original color image acquisition process of object, skewed slot Place plane is vertical with the camera plane of CCD color camera, and skewed slot is black with space wall locating for CCD color camera entirely Color.
Tealeaves recognition methods above-mentioned based on image procossing, in the RGB original color image acquisition process of object, CCD For the background of color camera using flannelette acrylic board as background board, light source uses colour temperature for the area source of 5000k-6000k.
Tealeaves recognition methods above-mentioned based on image procossing, in the RGB original color image acquisition process of object, recruitment The finished tea of camera imaging area is passed through in the shooting of industry line array CCD color camera, its pixel data is obtained, then by pixel data It carries out inputting tea dry sorting control system after processing is converted into RGB original color image.
Tealeaves recognition methods above-mentioned based on image procossing, the specific side of RGB original color image filtering and noise reduction pretreatment Method is to be filtered noise suppression preprocessing to RGB original color image using adaptive-filtering and histogram equalization method.
Tealeaves recognition methods above-mentioned based on image procossing, in the acquisition process of characteristics of objects parameter, the feature of object After parameter obtains, by Morphological scale-space, region ineligible in HSV color image is eliminated, and will be eligible Characteristics of image identify, then qualified image features are matched with database again.
Tealeaves recognition methods above-mentioned based on image procossing, after the RGB original color image of object obtains, by it It is input in tea dry sorting control system and is handled, extract the characteristic parameter based on HSV color image.
Tealeaves recognition methods above-mentioned based on image procossing, step C, in the matching process of characteristic parameter, the category of object Property confirmation after, tea dry sorting control system control one group high speed spray valve events, using high speed spray valve change yellow leaf, tea stalk and The direction of other materials landing, so that tealeaves, tea stalk, yellow leaf and other materials are landed from different channels.
Tealeaves recognition methods above-mentioned based on image procossing, the high speed spray valve and tealeaves feed chute are coplanar, position In the flat lower section in camera shooting area.
Beneficial effects of the present invention: compared with prior art, the present invention is based on the tealeaves recognition methods of image procossing, are mentioning Hsv color model is selected when taking tealeaves Color characteristics parameters, and extracts tealeaves, tea in color image H, S, V triple channel respectively Stalk, yellow leaf characteristic parameter, while determine tea stalk and yellow leaf H, S, V value range, by by the characteristic parameter of object be worth model Enclose and matched, thus by image multiple tea stalks and yellow leaf all identify, realize tealeaves identification automation, energy It is enough in real time in tealeaves tea stalk and yellow leaf be monitored, tealeaves will not be damaged in entire assorting room, Er Qiefen Select fast speed.It is identical using the COMPUTER DETECTION result that this method carries out with artificial detection result.
Detailed description of the invention
Fig. 1 is the color identification process figure of finished tea;
Fig. 2 be tealeaves, tea stalk, yellow leaf original image;
Fig. 3 is the channel HSV color image H pixel distribution;
Fig. 4 is the recognition result of tea stalk, yellow leaf.
Specific embodiment
The present invention is further illustrated with reference to the accompanying drawings and examples, but be not intended as to the present invention limit according to According to.
The embodiment of the present invention: a kind of tealeaves recognition methods based on image procossing, as shown in attached drawing 1-4, this method packet Include the foundation of the database of tealeaves, tea stalk and yellow leaf in known finished tea, the acquisition and characteristics of objects of characteristics of objects parameter The matching of parameter, specific steps include:
A, in known finished tea the database of tealeaves, tea stalk and yellow leaf foundation
(A1) in same kind of finished tea, the finished tea sample of multiple batches is chosen, should be wrapped in the finished tea sample Tealeaves, tea stalk and yellow leaf are included, and by all artificial separation of tealeaves, tea stalk and the yellow leaf in each sample, is individually stored, so The RGB original color image of tealeaves in each batch sample, tea stalk and yellow leaf is obtained afterwards;
(A2) numerous RGB original color images is made into filtering and noise reduction pretreatment respectively;
(A3) HSV color space will be transformed into from RGB color by pretreated RGB original color image form HSV Color image;
(A4) tealeaves, tea stalk and yellow leaf are extracted in the tone H of HSV color image, saturation degree S, these three channels lightness V Characteristic parameter;
(A5) according to the characteristic parameter of the tone H of tealeaves, tea stalk and yellow leaf, saturation degree S, lightness V, determine tealeaves, tea stalk with And the upper and lower limit value of the respective tone H of yellow leaf, saturation degree S, lightness V, so that it is determined that tealeaves, tea stalk and the respective color of yellow leaf Adjust the value range of H, saturation degree S, lightness V;
(A6) by the tea dry sorting control system of the value range typing tea dry sorting equipment of determining tealeaves, tea stalk and yellow leaf In, database is constituted, and saved.
After finished tea is based on the characteristic parameter Database of HSV color image, it can start to carry out based on the number According to the sub-sieve work of the finished tea in library.Sub-sieve work mainly includes several steps: firstly, being each in acquisition finished tea The characteristic parameter of a component part;Secondly, the characteristic parameter that will acquire is matched with the value range in database, the spy is confirmed Sign parameter is fallen into that value range, to confirm the attribute of the corresponding object of this feature parameter.So far, entire finished tea Identification process finish.
But the identification process of finished tea can only finally confirm in finished tea which be tealeaves, which be tea Stalk, which be yellow leaf, be other materials there are also those.The identification of finished tea is finally that the sorting to tealeaves provides service, most In whole assorting room, early-period confirmation is crossed to the finished tea of attribute by mechanical action, be divided into tealeaves, tea stalk, yellow leaf and Other four classifications, the substance of four classifications is collected together respectively.
B, in finished tea characteristics of objects parameter acquisition
Finished tea to sub-sieve is put into tea dry sorting equipment, in tealeaves transmission process, is repeated step (A1) to (A4), The characteristic parameter based on HSV color image of object is obtained by tea dry sorting control system;
C, the matching of characteristics of objects parameter
The value model of tealeaves, tea stalk in the characteristic parameter and database of the object that tea dry sorting control system will acquire and yellow leaf It encloses and is matched, confirm that each object belongs to tealeaves, tea stalk, yellow leaf or other.
In whole process, known to step A in finished tea the database of tealeaves, tea stalk and yellow leaf foundation, be based on people Work is by tealeaves, tea stalk and yellow leaf good sorting;And the matching process of step C characteristics of objects parameter, it is by machinery to made tea Each object in leaf is classified according to the difference of characteristic parameter.
It step B, in finished tea in the acquisition process of characteristics of objects parameter, include first step, as object RGB original color image obtains.And the RGB original color image acquisition methods of object are, finished tea is put in a vibration On device platform, as vibrator vibrates, the skewed slot even drop down that finished tea is 30 ° -70 ° along gradient, into CCD coloured silk The shooting area of form and aspect machine obtains RGB original color image by CCD color camera.After RGB original color image obtains, by it It is input in tea dry sorting control system and is handled, subsequent several steps can be completed by tea dry sorting control system, Extract the characteristic parameter based on HSV color image.
In the RGB original color image acquisition process of object, plane where skewed slot should be with the camera plane of CCD color camera Vertically, and space wall locating for skewed slot and CCD color camera is black entirely.
In the RGB original color image acquisition process of object, the background of CCD color camera uses the conduct of flannelette acrylic board Background board, light source use colour temperature for the area source of 5000k-6000k.
In the RGB original color image acquisition process of object, imaged with the shooting of industrial line array CCD color camera by camera The finished tea in region obtains its pixel data, then by pixel data carry out processing be converted into it is defeated after RGB original color image Enter tea dry sorting control system.
RGB original color image filtering and noise reduction pretreatment specific method is, using adaptive-filtering and histogram equalization Method is filtered noise suppression preprocessing to RGB original color image.
In the acquisition process of characteristics of objects parameter, after characteristic parameter obtains, tea dry sorting control system passes through form Processing, eliminates region ineligible in HSV color image, and qualified characteristics of image is identified, then Qualified image features are matched with database again.The size difference of each tealeaves will not in finished tea Very big, tea stalk is also to be compared by size in this way, be advantageous in that by Morphological scale-space there are also yellow leaf, can by it is some not It is that the part of tealeaves, tea stalk and yellow leaf directly weeds out, these substances directly weeded out do not need to carry out scheming based on HSV again The characteristic parameter of picture matches.For example, if tealeaves conveying or remove it is remote during, can generate some tealeaves powders, tealeaves powder it is big Small difference of comparing with tealeaves, tea stalk and yellow leaf is very big, this when passes through Morphological scale-space, it was found from the comparison i.e. of its size The road tealeaves powder is not belonging to any type in tealeaves, tea stalk and yellow leaf, to be divided into other one kind.This Sample, in the characteristic parameter matching process of subsequent progress, there is no need to carry out the characteristic parameter of the tealeaves powder and database It has matched, to reduce the workload of tea dry sorting control system.Realize the classification of tealeaves, tea stalk, yellow leaf and other materials It collects.
The high speed spray valve and tealeaves feed chute are coplanar, are located at the flat lower section in camera shooting area.Valve pair is sprayed using high speed In tealeaves, tea stalk, yellow leaf and other materials assorting room, it can include and take the problem of comparing out of.It is both to sort out tea stalk When, part tealeaves can inevitably be taken out of together.Under normal circumstances, recognizer is taken out of than meeting the requirements.If band Ratio is relatively low out, then can be neglected, can be on tea stalk, yellow leaf and other materials channel if taking out of than relatively high Industrial line array CCD color camera and high speed spray valve are set again, repeat the work of step B and C, until taking out of than lower than setting ratio Thus.And take out of than be it is how many, can be measured by multiple trial operation, or by tea dry sorting control system to CCD colour The image of camera shooting is analyzed to obtain.
Step A, in sample finished tea in the establishment process of the database of tealeaves, tea stalk and yellow leaf, and it is color using CCD Form and aspect machine is shot, and shooting environmental is identical as the shooting environmental in the acquisition process of step B characteristics of objects parameter.
Step C, in the matching process of characteristic parameter, after the attribute confirmation of object, tea dry sorting control system control one Group high speed spray valve events, using high speed spray valve change yellow leaf, tea stalk and other materials landing direction so that tealeaves, tea stalk, Yellow leaf and other materials are landed from different channels respectively.
For the finished tea of same type, only need early period to establish a database, and be saved to tea dry sorting control In system processed.Later period, then when needing to sort the finished tea of same type, that is, can skip this mistake of Database Journey, the characteristic parameter for directly carrying out unknown tealeaves obtains and the matching process of characteristic parameter.The finished product of so-called same type Tealeaves refers mainly to the tealeaves of the identical type produced using identical production method and its technique.

Claims (10)

1. a kind of tealeaves recognition methods based on image procossing, it is characterised in that: this method includes the foundation of database, object spy It levies the acquisition of parameter and the matching of characteristic parameter, specific steps includes:
A, the foundation of database
(A1) the RGB original color image of sample tealeaves, tea stalk and yellow leaf is obtained;
(A2) RGB original color image is made into filtering and noise reduction pretreatment;
(A3) HSV color space will be transformed into from RGB color by pretreated RGB original color image form HSV Color image;
(A4) the tone H of tealeaves, tea stalk and yellow leaf HSV color image, the characteristic parameter of saturation degree S, lightness V are extracted;
(A5) according to the characteristic parameter of the tone H of tealeaves, tea stalk and yellow leaf, saturation degree S, lightness V, respective tone is determined H, the upper and lower limit value of saturation degree S, lightness V, so that it is determined that the respectively value range of tone H, saturation degree S, lightness V;
(A6) by the tea dry sorting control system of determining value range typing tea dry sorting equipment, database is constituted;
B, the acquisition of characteristics of objects parameter
Tealeaves to sub-sieve is put into tea dry sorting equipment, in tealeaves transmission process, step (A1) is repeated to (A4), passes through The characteristic parameter of tea dry sorting control system acquisition object;
C, the matching of characteristic parameter
The characteristic parameter for the object that will acquire is matched with the value range of tealeaves, tea stalk and yellow leaf in database, is confirmed Object belongs to tealeaves, tea stalk, yellow leaf or other.
2. the tealeaves recognition methods according to claim 1 based on image procossing, it is characterised in that: the RGB of object is original Color image acquisition methods are that finished tea is put on a vibrator platform, as vibrator vibrates, finished tea along The skewed slot even drop down that gradient is 30 ° -70 ° obtains RGB by CCD color camera into the shooting area of CCD color camera Original color image.
3. the tealeaves recognition methods according to claim 2 based on image procossing, it is characterised in that: the RGB of object is original In color image acquisition process, plane where skewed slot is vertical with the camera plane of CCD color camera, and skewed slot and CCD colour phase Space wall locating for machine is black entirely.
4. the tealeaves recognition methods according to claim 2 based on image procossing, it is characterised in that: the RGB of object is original In color image acquisition process, the background of CCD color camera using flannelette acrylic board as background board, light source use colour temperature for The area source of 5000k-6000k.
5. the tealeaves recognition methods according to claim 2 based on image procossing, it is characterised in that: the RGB of object is original In color image acquisition process, the finished tea of camera imaging area is passed through with the shooting of industrial line array CCD color camera, obtains it Then pixel data carries out pixel data inputting tea dry sorting control system after processing is converted into RGB original color image.
6. the tealeaves recognition methods according to claim 1 based on image procossing, it is characterised in that: RGB original color figure It is as filtering and noise reduction pre-processes specific method, using adaptive-filtering and histogram equalization method to RGB original color image It is filtered noise suppression preprocessing.
7. the tealeaves recognition methods according to claim 1 based on image procossing, it is characterised in that: characteristics of objects parameter In acquisition process, after characteristic parameter obtains, by Morphological scale-space, area ineligible in HSV color image is eliminated Domain, and qualified characteristics of image is identified, then qualified image features and database are carried out again Matching.
8. the tealeaves recognition methods according to claim 2 based on image procossing, it is characterised in that: the RGB of object is original After color image obtains, it is input in tea dry sorting control system and is handled, extracted based on HSV color image Characteristic parameter.
9. the tealeaves recognition methods according to claim 1 based on image procossing, it is characterised in that: step C, characteristic parameter Matching process in, after the confirmation of the attribute of object, tea dry sorting control system controls one group of high speed spray valve events, utilizes high speed The direction that valve changes yellow leaf, tea stalk and other materials landing is sprayed, so that tealeaves, tea stalk, yellow leaf and other materials are from difference Channel landing.
10. the tealeaves recognition methods according to claim 9 based on image procossing, it is characterised in that: the high speed sprays valve It is coplanar with tealeaves feed chute, it is located at the flat lower section in camera shooting area.
CN201811245694.XA 2018-10-24 2018-10-24 Tealeaves recognition methods based on image procossing Pending CN109376257A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811245694.XA CN109376257A (en) 2018-10-24 2018-10-24 Tealeaves recognition methods based on image procossing

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811245694.XA CN109376257A (en) 2018-10-24 2018-10-24 Tealeaves recognition methods based on image procossing

Publications (1)

Publication Number Publication Date
CN109376257A true CN109376257A (en) 2019-02-22

Family

ID=65401248

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811245694.XA Pending CN109376257A (en) 2018-10-24 2018-10-24 Tealeaves recognition methods based on image procossing

Country Status (1)

Country Link
CN (1) CN109376257A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113477555A (en) * 2021-07-22 2021-10-08 西华大学 Fresh tea sorting machine based on image processing

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN202009669U (en) * 2011-03-17 2011-10-19 福建闽瑞茶业有限公司 Separating device for tea leaves and tea stalks
CN102521564A (en) * 2011-11-22 2012-06-27 常熟市董浜镇华进电器厂 Method for identifying tea leaves based on colors and shapes
CN102750680A (en) * 2012-06-30 2012-10-24 四川农业大学 Tea image enhancement method based on empirical mode decomposition and color space
CN104222327A (en) * 2013-06-14 2014-12-24 吴耿茂 Tea stalk-sorting and color-sorting all-in-one machine
CN207366997U (en) * 2017-09-25 2018-05-15 贵州省机电研究设计院 A kind of Multifunctional tea intelligence Grading System
CN108154195A (en) * 2018-01-19 2018-06-12 镇江思泊丽农业有限公司 Tealeaves recognition methods and the tealeaves sorting equipment using this method
CN108416814A (en) * 2018-02-08 2018-08-17 广州大学 Quick positioning and recognition methods and the system on a kind of pineapple head
CN108671534A (en) * 2018-05-21 2018-10-19 湖南大学 A kind of robot Chinese chess beginning pendulum chess method and system based on objective contour and framework characteristic

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN202009669U (en) * 2011-03-17 2011-10-19 福建闽瑞茶业有限公司 Separating device for tea leaves and tea stalks
CN102521564A (en) * 2011-11-22 2012-06-27 常熟市董浜镇华进电器厂 Method for identifying tea leaves based on colors and shapes
CN102750680A (en) * 2012-06-30 2012-10-24 四川农业大学 Tea image enhancement method based on empirical mode decomposition and color space
CN104222327A (en) * 2013-06-14 2014-12-24 吴耿茂 Tea stalk-sorting and color-sorting all-in-one machine
CN207366997U (en) * 2017-09-25 2018-05-15 贵州省机电研究设计院 A kind of Multifunctional tea intelligence Grading System
CN108154195A (en) * 2018-01-19 2018-06-12 镇江思泊丽农业有限公司 Tealeaves recognition methods and the tealeaves sorting equipment using this method
CN108416814A (en) * 2018-02-08 2018-08-17 广州大学 Quick positioning and recognition methods and the system on a kind of pineapple head
CN108671534A (en) * 2018-05-21 2018-10-19 湖南大学 A kind of robot Chinese chess beginning pendulum chess method and system based on objective contour and framework characteristic

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
吴哲、刘孝星、郑力新、周凯: "《一种L-M优化BP网络的茶叶茶梗分类方法》", 《计算机技术与发展》 *
姚波、汪洋、李小瑞、吕军: "《自然环境下嫩芽与茶梗自动分割方法的研究》", 《黑龙江八一农垦大学学报》 *
黄留锁,宋艳: "《茶叶茶梗的计算机在线识别技术及其应用》", 《福建茶叶》 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113477555A (en) * 2021-07-22 2021-10-08 西华大学 Fresh tea sorting machine based on image processing

Similar Documents

Publication Publication Date Title
CN107016405B (en) A kind of pest image classification method based on classification prediction convolutional neural networks
EP2548147B1 (en) Method to recognize and classify a bare-root plant
CN103345617B (en) Chinese medicine knows method for distinguishing and system thereof
CN107330892A (en) A kind of sunflower disease recognition method based on random forest method
CN106997475B (en) A kind of pest image-recognizing method based on parallel-convolution neural network
CN102915446A (en) Plant disease and pest detection method based on SVM (support vector machine) learning
CN103323455A (en) Tobacco leaf grading method based on reflection, perspective and microscopic images
CN104198324B (en) Computer vision-based method for measuring proportion of cut leaves in cut tobacco
CN103514452B (en) A kind of fruit shape detection method and device
Sahu et al. Identification and classification of mango fruits using image processing
CN107978110A (en) Fence intelligence identifying system in place and recognition methods based on images match
CN107464035A (en) Chinese medicine performance rating method and system
Amlekar et al. Leaf features based plant classification using artificial neural network
KR20200084940A (en) CNN(Convolutional Neural Network) based pest and damage fruit classification device and method
WO2012172073A1 (en) Method for providing a prediction model for crack detection and method for crack detection on a semiconductor structure
Patki et al. Cotton leaf disease detection & classification using multi SVM
CN107121436A (en) The Intelligent detecting method and identification device of a kind of silicon material quality
CN103090946B (en) Method and system for measuring single fruit tree yield
CN109376257A (en) Tealeaves recognition methods based on image procossing
CN109685071A (en) Brain electricity classification method based on the study of common space pattern feature width
CN108898156A (en) A kind of green green pepper recognition methods based on high spectrum image
CN201041547Y (en) Device for analyzing and recognizing different types of green teas based on multi-spectrum image texture
CN111191510A (en) Relation network-based remote sensing image small sample target identification method in complex scene
CN109815945A (en) A kind of respiratory tract inspection result interpreting system and method based on image recognition
CN109447009A (en) Hyperspectral image classification method based on subspace nuclear norm regularized regression model

Legal Events

Date Code Title Description
RJ01 Rejection of invention patent application after publication

Application publication date: 20190222